Mutations in the gene encoding methyl CpG-binding protein 2 (MECP2) cause Rett syndrome, a severe post- natal neurological disease, marked by developmental regression, autism-like features, seizures, and motor abnormalities. Duplications of MECP2 cause MECP2 duplication syndrome, which shares a number of features with Rett syndrome, including autism and epilepsy, and is a leading cause of X-linked intellectual disability. Animal studies show that the brain is extremely sensitive to MeCP2 protein levels such that a twofold change up or down causes neuropsychiatric disease, suggesting that perturbations of MeCP2 modulators could cause disease by dysregulating MeCP2. However, little is known about MeCP2 regulation. An unbiased siRNA screen of polyadenylation factors showed that NUDT21 regulates MECP2 alternative polyadenylation and protein levels in vitro. Thus, I hypothesize that NUDT21 regulates MeCP2 in vivo, and that its loss or gain causes neuropsychiatric disease, partly by altering MeCP2 levels. To this end I gathered preliminary data from neuropsychiatric patients with copy-number variations (CNVs) spanning NUDT21, and showed that NUDT21- encoded CFIm25 and MeCP2 protein levels inversely correlate. Moreover, when CFIm25 levels are normalized in duplication patients by anti-NUDT21 siRNA, MeCP2 levels are restored, showing that NUDT21 alone regulates MeCP2 levels. I also found MECP2 mRNA increases in both the NUDT21 deletion and duplications patients, but that the increase in the deletion patient comes from an increase in short MECP2 isoforms, whereas the elevated mRNA from the duplication patients comes largely from long MECP2 isoforms, which harbor known regulatory miRNA binding sites, and are inefficiently translated. To confirm that perturbed NUDT21 alters MeCP2 levels and causes the phenotypes in patients with the CNVs, I have generated Nudt21 knockout mice, and have acquired a mouse line with a Nudt21 conditional allele. I will characterize their behavioral, cellular, and molecular phenotypes, including RNA-seq to assess expression and alternative poly- adenylation changes. To determine the degree that the phenotypes of patients with CNVs spanning NUDT21 are driven by MeCP2 dysregulation, I will cross Nudt21 heterozygous mice with mice expressing a Mecp2 allele that causes 50% reduction of MeCP2 and display neurobehavioral problems. In the double mutant offspring, I can determine if normalizing MeCP2 levels will rescue some phenotypes of the Nudt21 heterozygous mice and ascertain the contribution of MeCP2 dysregulation to their phenotypes. This will demonstrate the feasibility of using MeCP2 reducing drugs now in development for the treatment of NUDT21 loss-of-function phenotypes. This proposal will test the hypothesis that NUDT21 loss of function causes disease that NUDT21 regulates MeCP2 in vivo, and the degree to which MeCP2 dysregulation drives the Nudt21 loss-of-function phenotypes. The results will potentially provide some neuropsychiatric patients with a diagnosis, and direct future treatment strategies.